Reference implementations for using Kubernetes-based Event Driven Autoscaling (KEDA) for scaling containerized workloads hosted on AKS cluster.
Kubernetes does resource-based scaling on metrics defined by CPU and memory really well. Event-driven applications require a more targeted solution to scale proactively. With KEDA, you can drive the scaling of any workload in Kubernetes based on the number of events/requests needing to be processed.
KEDA is a single-purpose and lightweight component that can be added into any Kubernetes cluster. KEDA is an open source, CNCF solution. It works alongside standard Kubernetes components like the Horizontal Pod Autoscaler (HPA) and Cluster Autoscaler, and can extend its functionality. With KEDA, you can explicitly target the apps you want to use event-driven scale, with other apps continuing to function using other scaling methods. This makes KEDA a flexible and safe option to run alongside any number of any other Kubernetes applications or frameworks.
KEDA works alongside standard Kubernetes components like the Horizontal Pod Autoscaler (HPA), and the Cluster Autoscaler. KEDA extends auto-scaling functionality without overwriting or duplication.
The diagram below shows how KEDA integrates with the Kubernetes HPA, external event sources, and Kubernetes’ API Server to provide autoscaling functionality.
Read about how KEDA works and more in depth details from their official documentation.